This Competitive Revision aims to greatly enhance our ability to estimate lifetime and age-specific solar ultraviolet (UV) exposures obtained retrospectively from self-report and enhance the novel UV exposure metric that we developed for, and are using in, the Parent Award. As well as enhancing the scientific approach of the Parent Award, this additional methodological work will have wide applicability to a number of other studies of UV exposure and disease outcome that have been initiated since our Parent Award began, by collaborators of the investigative team. We previously developed a method of imputing lifetime and age-specific UV exposure. We now have an opportunity to validate that UV metric against real-time UV dosimetry, and use that information to better measure retrospective UV exposure, especially extreme and intermittent exposures that are relevant to melanoma but not well measured by current methods. In the proposed study, we aim to characterize the relationship between imputed UV exposure and real-time personal UV exposure: that relationship is certainly mediated by sun exposure behavior (not all people will receive the same UV exposure just because they are in the same geographical location). For melanoma, a critical issue is the impact of past intermittent exposure on disease. Currently, intermittent UV exposure is measured by self-report of activities such as vacations in the sun, sun bathing, or short term weekend exposures. These self-reports almost certainly misclassify intermittent exposure estimates, and it is unclear which measure best represents intermittent UV exposure. Our UV metric measures total UV, but for melanoma we also want to know the component of total UV exposure that is attributable to specific wavelengths (crudely, UVA and UVB). Our dosimeters have been developed in order to correlate wavelength-specific UV measures with self-reported sun exposure behavior. We will measure UV from dosimetry in the Parent Award study population, and use those data to model the relationship between self-reported measures, our imputed UV metric (both of which are available for retrospective data also) and real time UV exposure (especially intermittent and extreme exposures, and the contribution of UVA and UVB wavelengths separately). The proposed study will further accelerate our progress towards the central theme of reducing exposure assessment error in order to determine the true effects of UV exposure on melanoma. This Competitive Revision application meets ALL of the criteria for funding under the ARRA, both for accelerating the tempo of scientific research directly for the Parent Grant and for associated science, and for job creation/retention that is the primary focus of the Stimulus Bill. PUBLIC HEALTH RELEVANCE: Melanoma incidence is increasing steadily in most populations worldwide, and primary prevention efforts (largely based on avoiding sun exposure around midday) have had limited success. Our Parent Award aims to better characterize the role of varying wavelengths of solar ultraviolet (UV) on melanoma, in the absence of the muddying effect of genetic susceptibility, and provide the 'cleanest'picture of how to achieve primary prevention through behavior change. The proposed study enhances those aims by relating imputed UV exposures (based on self-report) to real time UV exposure from dosimetry, using devices that also measure wavelength-specific UV. Those dosimetry data will be modeled against self report and imputed UV data to determine the best way to measure variability in past UV exposures that is relevant to melanoma, focusing specifically on intermittent and extreme exposures that current methods almost certainly fail to accurately capture.